Book Review: International Courts and the Performance of International Agreements

Compliance with international agreements is among the most debated topics in legal literature. By and large it is assumed that international courts have no direct control over national governments, but can nevertheless facilitate cooperation, albeit with some important political constraints. Building upon relatively recent studies, Carrubba and Gabel propose a formal theory of international court influence and test it against the European Union, as a regional, sui generis, international organization. After a general introduction regarding the ongoing debate on international agreements and courts from a liberal institutional perspective and a rational design perspective, in Chapter 1 the authors present their theoretical approach. Most notably, they “want a model that is general (i.e., not tailored to a specific regime) that explicitly endogenizes each step of the adjudication process” (p. 26). Regardless of the specific regulatory regime created under the relevant international agreement (trade, environment, human rights, etc.) national governments must be driven by the need to pursue a collective action. In this sense, international courts contribute to upholding the law and ensuring cooperation, but noncompliance is believed to be physiological and accepted to a certain extent, even in national (federal) systems.

Chapter 2 clarifies the theory put forward by Carrubba and Gabel.Their model rests on the assumption that: "cooperation is sustainable as long as the long-run benefits from cooperation exceed the short-run cost of complying" (p.31) And yet: "deeper agreements lead to more variable costs, and the more variable the costs, the more likely a government will face a situation when the short-run cost of compliance is simply too severe, and the government will defect, even knowing that doing so will lead to punishment at a minimum and the end of cooperation entirely in the worst-case scenario " (p. 31).This also explains why exceptions are normally included in these agreements.On the other hand, only individual governments are able to determine the actual costs of compliance.Thus, the need for an institutionalized court.In this regard, the dispute-generation and -resolution processes in place are considered pivotal.While the former must be designed to allow for effective standing to those actors that can be hindered by the relevant breach, the latter must ensure that third-party governments may intervene.Especially in advanced multilateral treaties (e.g., trade) that provide for adequate procedures and third-party participation, the competent courts can act as a fire alarm and information clearinghouse.International courts tend to "maximize government compliance with adverse rulings" (p.38), but at the same time bear a legitimacy cost if their rulings are not observed.In light of the above, the authors argue, firstly, that "the court is more likely to rule against a defendant government the more briefs filed against the government and the fewer briefs filed in support of it" (p.47) (the Political Sensitivity Hypothesis) and, secondly, that "court rulings against defendant governments are more likely to change government behavior the more briefs filed in support of the ruling and the fewer filed against it (the Conditional Effectiveness Hypothesis) (p.47).Of course, this does not prevent international courts from innovating the existent regulatory regime; it simply implies that progress will be possible only insofar as national governments do not perceive them as mutually costly.

Testing the Theory in the European Union
In Chapter 3, Carruba and Gabel test the viability of their two main propositions in the European Union.They do so by analyzing an original data set covering all the judgments of the Court of Justice in annulment and infringement proceedings, as well as in preliminary references, between 1960 and 1999.The EU legal order appears to be particularly well suited: it pursues economic integration; it foresees a central dispute settlement mechanism administered by the Court of Justice; it recognizes standing to institutions, Member States and, albeit to a limited extent, individuals, and it offers third-party governments the option to intervene in support of or against the defendant or the plaintiff.Moreover notwithstanding the lack of an autonomous enforcement mechanism, the Court of Justice (ECJ) is known to follow an 'integrationist agenda', which increases the risk of noncompliance.The authors thoroughly examine the distribution of litigants in annulment proceedings and preliminary rulings; indicate the number of annulment and infringement rulings by decade (60s, 70s, 80s and 90s); offer a breakdown of the preliminary rulings by the referring nation and decade; capture the distribution of third-party briefs by country and decade and reveal the percentage of favorable rulings for the plaintiff.The risk of noncompliance is common to all three types of actions under scrutiny and the gathered data suggests that there is indeed a relation between third-party government briefs (in support or against the defendant or plaintiff) and the finding of the Court.
In Chapter 4 the authors measure the legal quality of the arguments based on the analysis of the Advocate General (AG) (as opposed to those advanced by the parties and interveners) in order to discard the possibility that the Court actually rules according to the balance of the legal merits in the case.The AG, in fact, is believed to offer all the relevant guarantees: legal expertise and motivation to assess the case on the merits, and the merits alone.Of course, political influence cannot be excluded, but the instances in which the opinion can be said to be 'suspect' are very limited.The analysis of the relevant data confirms that there is a wide correspondence between the opinions and the judgments by the ECJ and, by consequence, the position of the AG does in fact reflect the legal merits of a case.
The prediction that a ruling against a defendant government is related to the number of (net) briefs submitted against it (the Political Sensitivity Hypothesis) is measured and tested in Chapter 5.The baseline analysis covers an impressive 1,098 rulings with a defendant government and 1,808 rulings with a litigant government.The first set of rulings is tested more narrowly (taking into consideration rulings in which the government was a defendant), whilst the second set of rulings is tested more broadly (taking into consideration also annulment proceedings and preliminary rulings in which the government was a plaintiff).In both cases the results are consistent with the expectations, although the Court appeared to be "much more sensitive to government briefs when governments were litigants (when threats of noncompliance would be most present) than otherwise" (p.155).
The prediction that "European Court of Justice rulings against governments are more likely to change the behavior (i.e., change policy) when the briefs filed by third-party governments are (net) against the government" (p.156), in turn, is gauged in Chapter 6 by analyzing how the ECJ rulings impact on EU interstate commerce.The authors examine preliminary rulings and infringement rulings, distinguishing them based on the risk of noncompliance.They estimate their separate and combined effect on trade in the six original Member States from 1970 to 1993 and consider both their short and long-run effects on intra-EU imports, weighting third-party briefs in support of the plaintiff in light of the political and economic importance of the interested country calculated through logged GDP.The analysis shows, "that tradeliberalizing rulings that attract the net support of third-party government briefs should be more likely to enjoy government compliance and thereby have a policy impact than would trade-liberalizing rulings that lack such third-party support" (p.188).And interestingly enough, the major difference between preliminary rulings and infringement rulings appears to be one of quantity as the former seem to have a greater impact.On the other hand, intra EU-imports in the Member States are not affected in the absence of net third-party government supporting briefs.
In their concluding remarks, Carrubba and Gabel contest the (still) predominant neofunctionalistic view that integration occurs independently of government preferences.Resuming the most salient aspects of the investigation, the authors suggest that: "we need to move past asking whether international institutions are effective at changing government behavior to asking under what conditions international institutions are effective" (p.194).Indeed, their findings indicate that the Court is responsive, and more deferential, to governments' positions when noncompliance is perceived as a realistic consequence of a ruling.

Final Remarks
Through formal modeling, or game-theory, the book offers a fascinating portrayal of the dynamics that govern judicial adjudication vis à vis market integration in the European Union.The methodological rigor that characterizes the work -in particular the consistent search for alternative explanations -supports the validity of the theory and stimulates further reflection on its practical implications.The tables reflecting the extensive and in-depth study favor clarity and accessibility by students, academics and practitioners, even those not familiar with regressive analysis.
Nonetheless, perhaps because of their background, the authors appear to overlook a number of events and legal developments that occurred after 1999 that could impact on the reliability, today, of results of their research.Firstly, the reform of the Statute approved on the occasion of the signature of the Nice Treaty cases can be (and routinely are) decided without the contribution of the AG, which makes it now impossible to measure the legal quality of the arguments based on the analysis of the opinion, as opposed to those advanced by the parties and interveners.Secondly, the 2004 and 2007 enlargements have deeply affected the functioning of the Court and the weight of, and balance between, Member States.Thirdly, the Lisbon Treaty has empowered the Commission to immediately request the payment of a penalty whenever a Member State fails to notify measures transposing a directive adopted under a legislative procedure (Art.260(3) TFEU).This inevitably enhances the deterrent effect of infringement actions and possibly impinges on the soundness of the theoretical model.That being said, the book by Carrubba and Gabel remains a well-conceived and intriguing volume on the role of the Court of Justice in the European integration process.Indeed, despite the many contributions to legal literature on the adjudication process and the impact of its rulings in the Member States, none address the role of the Luxembourg judges pursuant to the predictions posited by the authors.The complexity of the topic and the innovative approach followed by the authors make this volume a valuable learning instrument for graduate students of political sciences and law, in particular PhD candidates working with regressive analysis, but also for judges and lawyers interested in the enforcement of international agreements.